On computing decision regions with neural nets

‘ Ah Introduction to Computing with Neural Nets

Artificial neural net models have been studied for many yeara in the hope of achieving human-like performance in the fields of speech and image recognition. These models are composed of many nonlinear computational elements operating in parallel and arranged in patterns reminiscent of biological neural nets. Computational elements or nodes are connected via weights that are typically adapted du...

Computing on Analog Neural Nets with Arbitrary Real Weights

The Applicability of Neural Nets for Decision

The aim of this paper is to discuss the possible role of neural nets for decision support. The discussion will be conducted along two connected lines. The rst line regards the possibilities to solve combinatorial optimization problems with multi-layered perceptrons. Particular attention is paid to the required complexity of such neural nets. The second line regards the use of neural nets as a k...

Decision trees, boosted decision trees, branching programs, neural nets intro

Administrative: • Enrollment should be good now. • Next homework will go out next Wednesday or Friday. • Recap of where we are in course, and interplay between representation, optimization, and generalization; usual story is that as representation power improves, optimization and generalization become more painful. • Many basic latex errors in assignments; please consult lshort for an easy refe...

Computability with Neural Nets

This paper shows the existence of a nite neural network, made up of sigmoidal neurons, which simulates a universal Turing machine. It is composed of less than 10 5 synchronously evolving processors, interconnected linearly. High-order connections are not required.